Development of a Collaborative Framework for Efficient Calibration of LRFD Geotechnical Resistance Factors and Data Exchange
The AASHTO LRFD design code allows for flexibility in geotechnical resistance factors applied within its design framework. While most State Departments of Transportation use the ‘default’ geotechnical resistance factors provided in the national code, many of which come from the early ASD fitting or national studies, some states have invested in developing more accurate locally calibrated factors. Based on reliability and statistical analysis of predictive information from local geology, exploration methods, design and construction practices, and associated performance measurements, these resistance factors, establish more representative values with reduced variance for local conditions. FHWA Report FHWA NHI-10-039, “Implementation of LRFD Geotechnical Design for Bridge Foundations” describes implementation of the LRFD design specifications with consideration of their local experiences. Local reliability calibrations provide a mechanism to capture value and promote/justify the consideration of better methods prediction methods and performance verification, leading to larger resistance factors, φ, and improved efficiency and economy while preserving an appropriate level of risk and reliability.
Because these practice improvements require a sound statistical basis, there can be a large investment cost. An obstruction to quicker practice improvement is inefficient data management. The focus of this work is to improve foundation design and construction practices through improved, standardized, asset management.
Additional research is needed to accelerate the development of a complete framework for the systematic, repeatable, and standardized acquisition of data necessary for the entire calibration process. This process includes a taxonomy [classification system ] associated with the quality of preconstruction site assessment, construction performance measurements, and a common data exchange protocol such that as more information is added datasets can be re-evaluated and efficiently expanded and readily incorporated into calibration efforts on local and regional scales.
Development for the expansion of the DIGGS XML interchange schema associated with this work will harness good data management practices for improved simplicity in data exchange and consistency in the collected information, improving data quality and providing improved potential for interoperability. The nature of DIGGS XML which carries geospatial data will provide extraordinary value for assessment of site variability using geostatistical models and provide datasets for artificial intelligence and machine learning.
The objectives of the proposed research are to provide system architecture and supporting process elements for efficient and repeatable local calibration of geotechnical resistance factors. [Once those factors are calibrated, improved site characterization and/or testing will lead to more favorable resistance factors]. Examples of different process components and processes that must be developed to accomplish the project’s objectives include:
Calibration requires initial site characterization information in combination with foundation performance testing for a complete record of both predicted and measured performance of foundation systems. A primary objective is to provide a taxonomy associated with the data quality by defining classes [categories, grades, levels] of supporting information, and the required elements associated with each class, to be collected as 1) predictive characterization, 2) project metadata, and 3) construction performance monitoring. The categories are associated with various levels of effort, quality, or completeness and the overall value and quality of the data for the site.
Develop documentation and processes to ensure necessary information related to site variability is included. As part of this objective, the benefits associated with additional testing, better quality testing, more diverse testing [multiple independent methods] will be identified and documented, establishing a link between better predicative data and the ability to derive and later use improved resistance factors. An outcome would be to provide different resistance factors based on the site investigation class [category, grade, level]
Develop data collection and input guidance and note responsibility areas and suggested standards of practice and care;
Develop a data exchange structure within the DIGGS XML schema to enable the common and efficient exchange of information among those collecting the information and those developing tools to use the information for LRFD calibration.
Define required site metadata and measurement information to be collected as performance monitoring for constructed foundation elements or foundation systems (strength or service load tests);
Provide necessary process documentation such that future studies can efficiently use this framework [process] to evaluate predictive measurements and load testing measurements to site variability. The classes of data quality will allow DOTs to select the level of effort appropriate for sites and apply the associated resistance factors. Improved operational processes will result in the ability to use resistance factors associated with that level of effort.
Detail the complete collaborative calibration process from ‘cradle to grave’ such that this process can be efficiently specified and used by those parties involved in the local calibration process and allow the efficient creation, use, and update of applications using the collected data;
Provide language to DOTs for use in contract specifications and requirements documents. Assist states in developing policy to require the use of the framework when performing work useful to LRFD calibration or validation.
This research project will provide a system for data collection, structure, and transmission of geotechnical information needed for calibrating LRFD resistance factors for shallow and deep foundation systems. With a defined nomenclature, taxonomy [classification system for quality] procedure, and data structure, a repeatable and efficient procedure will be established for local calibration of resistance factors.
This work will translate to shortened project durations, reduced project expenses, and either reduced risks or more well-defined risks, reducing unnecessary or inappropriate conservatism. Development of improved resistance factors provides a direct and repeatable opportunity ability to apply more favorable resistance factors. Those factors can be researched, developed, and implemented easier and faster with the products delivered from this work.
The long-term benefits of this standard data structure and improved ability to exchange geotechnical LRFD calibration data includes:
Creating codified incentives for improved exploration, design, or performance evaluation improving economy or safety through comprehensive and consistent design considerations of variability and performance.
Providing a system to potentially significantly decrease the effort in developing and deploying geotechnical strength resistance factors which are based on parameters developed at the time of design with respect to local considerations.
Providing a system to efficiently allow calibration and valuation of local service limit geotechnical resistance factors. Adding to the utility of the DIGGS XML data interchange standard and the return on investment for developing applications which use the standard, proving greater economy to developers and DOTs using these tools. The ability to share and use data in a standardized system across the national system in addition to just localized use increases the value of the project information. As local factors are developed, national factors could be developed and revised as additional data is gathered. Project information could easily be plotted on national maps to improve understanding of the spatial coverage of the information. Subsequent testing (of any kind) could be better targeted to address knowledge gaps observed in the data rather than just performing a test to “do a test.” Economies of scale for time and money aspects of additional project work; where information already exits, using this framework, additional efforts may be reduced, which could benefit both traditional and alternative procurement projects.
Improving calibrations for the geotechnical strength and service limit states based on full scale load testing and assessment of the performance of in-service structures improving economy and efficiency by reducing unnecessary or inappropriate “extra” conservatism.
Establishing preferred information for levels of quality or effort for all the phases of the calibration process as well as the data interchange schema, enabling the workflow to be improved and standardized, allowing efficient development and deployment of applications. Capturing high quality data in a consistent, easily sharable, format will empower DOTs, Academics, and Industry Professionals to reliably build multi-use interpretive models that will enhance design methods, improve quality and reliability, and establish performance baseline data for geotechnical assets where significant national investments are made annually.
Allowing levels of effort or quality to be examined as components from robust datasets such that reliability and risk relationships are developed based on site conditions, variability, and the amount and quality of geotechnical site characterization and performance monitoring applied.
Components of the work can be used to help promote other initiatives such as 1) development of quality source files for use in artificial intelligence and neural network studies; 2) exploring the quality afforded by direct CPT design for prediction of Bridge Foundation design and performance at the geotechnical service and strength limit states; 3) assessment of site characterization quality and variability and the impact of uncertainties in bridge design and performance; 4) examining geotechnical instrumentation and bridge foundation design and performance; 5) acquiring consistent and quality data for use in examining service, strength, and extreme event limit state design more generally.
Allen, T. (2005). “Development of Geotechnical Resistance Factors and Downdrag Load Factors for LRFD Foundation Strength Limit State Design.” Federal Highway Administration, Washington, D.C., Publication No. FHWA-NHI-05-052. http://www.wsdot.wa.gov/biz/mats/Geotech/ResistanceFactorsforFoundationsFinal2_05. pdf
Allen, T.M., Nowak, A.S., and Bathurst, R.J. (2005). “Calibration to Determine Load and Resistance Factors for Geotechnical and Structural Design.” Transportation Research Circular E-C079, Transportation Research Board, Washington, D.C., 83 p. http://onlinepubs.trb.org/onlinepubs/circulars/ec079.pdf
Duncan, J.M. (2000). “Factors of Safety and Reliability in Geotechnical Engineering.” Journal of Geotechnical and Geoenvironmental Engineering, Vol. 126, No. 4, pp. 307-316.
Paikowsky, S.G., Birgisson, B., McVay, M., Nguyen, T., Kuo, C., Baecher, G., Ayyub, B., Stenersen, K., O’Malley, K., Chernauskas, L., and O’Neill, M. (2004). “Load and Resistance Factor Design (LRFD) for Deep Foundations.” NCHRP Report 507, Transportation Research Board, Washington, D.C., 126 pp. http://www.trb.org/news/blurb_detail.asp?id=4074
Samtani, N.C., and Nowatzki, E.A. (2006). “Soils and Foundations: Reference Manual.” National Highway Institute, Federal Highway Administration, Washington, D.C., Publication No. FHWA-NHI-06-088, 462 p.
The proposed research consists of the following tasks:
Task 1: Review Literature and Assess Current Practices
This task includes a review of national practices for geotechnical LRFD calibration and current data formatting and how those practices can be adapted for improved development moving forward with best practices for site characterization and performance monitoring.
Task 2: Develop a Taxonomy Related to Quality of Site Investigations [Minimum Requirements, Elements, or Performance-based Metrics]
Perform appropriate studies to create or document a new taxonomy [classification scheme] associated with site investigation practice and levels of quality. Quality would be based on considerations including amount of sampling, methods employed, if multiple methods are employed, etc. Review existing literature and guidance highlighting clarifying elements for the capture of high-quality data. Assessment of site variability, risk, and accuracy, precision, repeatability, and other sources of error for measurement values with respect to predictive estimation of future foundation performance would also be evaluated. Classification/codification would be similar to distinctions made in construction control ‘quality’ [practice and methods used and how many samples] for driven pile construction control.
Developing a set of uniform and accepted guidelines about what types of information should [or must] be captured prior to performance tests (site exploration, stratigraphy, material properties, ground water etc.) during the performance tests (execution and procedure) and when processing (analysis) as well as other test metadata (location, date, time after installation, element installation construction details). These could be ‘researched’ and developed with an eye to what information is necessary for robust statistical evaluations needed for use in LRFD resistance factor calibration.
Task 3: Develop Procedures and Guidance Related to Site Variability
Determine the elements of statistics-based decision making which need to be introduced into the process. Address project planning and decision-making with respect to site variability and principles such as variance, standard deviation, COV, and geo-statiscical evaluation.
Task 4: Develop Necessary DIGGS XML schema elements.
Perform appropriate studies development to expand the DIGGS schema to include necessary information for each of the phases of LRFD calibration information and metadata to support calibration efforts, applications, software, and statistical studies.
Develop as format (schema) for such a data set within the framework of the FHWA developed DIGGS structure so that future compilation of data sets can be transferred between users without proprietary or personal preference for the data structure. Where valuable high-quality data is archived, this could be translated into this DIGGS format for future use by industry.
Provide a DOT use case: These shallow and deep foundation load tests (piles, nails, anchors, shafts, footings, etc.) data sets could then be readily produced by contractors and delivered to DOTs and others by contract requirement using the DIGGS schema The data files could then be archived by the agency, imported into in house or commercial tools for interpretation or reporting, uploaded into central databases maintained by the DOT and readily retrieved and transmitted to consultants and researchers who have need cases or express freedom of information requests for such data.
Explain the Benefits: Once a DOT receives such a data set, it would be readily input into existing or new State DOT templates, Excel spreadsheet tools or archival data management systems. Retrieval and distribution of the data also would be nearly automated in this digital format.
Task 5: Develop the Data Management and Delivery Guidance for Foundations [This could be extended to transportation earthworks, if performance for those systems exists i.e. column supported embankments]
Develop a comprehensive “cradle to grave” best-practice framework for State DOTs and others to follow as a roadmap to acquire the necessary and complete data elements in the field at the time of capture to provide an appropriate basis for LRFD resistance factor statistical development, quality control and assurance as well as long term asset management.
Expand the Benefits to the Future: The load testing data (and supporting information) could be readily shared and searched internally within the DOT or shared with researchers or consultants for future interpretation and analysis in current or follow-on use in whatever Apps, widgets, and tools that may be available at the time. Agencies would collect their own information and share across agencies or with outside groups (DFI, PDCA, others) that may create data-warehouse tools of openly available data set. Open access would allow ready combination of such public data sets supplemented with private data sets for research/academic, industry advancement or other uses. Data in consistent formats would provide significant utility for use in artificial intelligence and other machine learning environments.
This research will serve as an important area for expansion and national implementation for the DIGGS XML interchange format and basis for development of applications supporting the continued evolution of the LRFD framework for both strength and service limit state factors for local and regional efforts to improve economy and safety by harmonizing the risk and reliability associated with construction of transportation structures. This would further provide reliable data sets for performance measurements and asset management.
This RNS, focusing on uniform process and systems supports harmonizing data and will make data readily accessible saving money and time on compilation of data, assembling for interpretation and research, need for redundant testing and inefficient or faulty design. This would also allow users/agencies to easily add data to their unique databases developed specifically to meet their unique needs.
This proposal is confined to developing tools such as the definitions, taxonomy, critical elements (requirements), process documentation, and data transfer protocols. It allows participants to benefit from a standardized system such that all contributed data has greater value and applications can be developed more efficiently.
Federal and State DOTs, Researchers; representatives from the following departments have indicated this RNS is an important area of development.
|Sponsoring Committee:||AKG70, Foundations of Bridges and Other Structures
|Research Period:||24 - 36 months|
|RNS Developer:||Derrick Dasenbrock, FHWA (and friends)|
|Source Info:||Cosponsoring Committees|
AKG60 Standing Committee on Geotechnical Instrumentation and Modeling
AKG20 Standing Committee on Soil and Rock and Site Characterization
|Index Terms:||Load and resistance factor design, Geotechnical engineering, Data communications, Data sharing, Calibration, |
|Cosponsoring Committees:||AKG60, Geotechnical Instrumentation and Modeling; AKG20, Soil and Rock Properties and Site Characterization|
Bridges and other structures